The impact of income, public assistance and homelessness on seeking medical care.
Objective: To assess gender differences and determine whether
economic determinants are correlated to delayed medical care among this
Methods: A survey was administered to 215 Houston residents recruited from clinics and club/ bars locally in 2005. Descriptive statistics and a logistic regression were utilized. Results: Participants with a history of homelessness (OR = 2.21), particularly women (OR = 3.03), men with income (OR = .09), and men with public assistance (OR = 16.33) were more likely to delay medical care.
Discussion: Gender differences between economic determinants and willingness to sustain physical health exists. Future interventions should aim to improve medical care access in spite of economic pitfalls.
Health care industry
Social service (Surveys)
|Publication:||Name: American Journal of Health Studies Publisher: American Journal of Health Studies Audience: Professional Format: Magazine/Journal Subject: Health Copyright: COPYRIGHT 2011 American Journal of Health Studies ISSN: 1090-0500|
|Issue:||Date: Summer, 2011 Source Volume: 26 Source Issue: 3|
|Topic:||Computer Subject: Health care industry|
|Product:||Product Code: 9105130 Social Service Support Programs NAICS Code: 92313 Administration of Human Resource Programs (except Education, Public Health, and Veterans' Affairs Programs) SIC Code: 8000 HEALTH SERVICES|
|Organization:||Organization: University of Texas|
|Geographic:||Geographic Scope: Texas Geographic Code: 1U7TX Texas|
Mortality and minimal survival rates disproportionately impact African Americans above any other ethnic or racial group in the United States (American Cancer Society, 2009). Distinctly different from other American nationalities, African Americans not only lead in health disparities, but also lead in three major areas: homelessness, unemployment, and public assistance use. Firstly, homeless families comprise 41% of the homeless population; however, African Americans make up 42% of that population (National Coalition for the Homeless, 2009). Although African Americans make up a mere 11% of the general population, this minority group represents 40% of homeless Americans (Public Broadcasting Network, 2007; National Coalition for the Homeless, 2009). Amidst a national recession, 43.6 million Americans live in poverty; however, African Americans are still experiencing poverty to a much larger degree compared to Caucasian counterparts (DeNavas-Walt, 2010). According to national data, African Americans have had the lowest median household in comparison to other races on average from 1967-2009 (DeNavas-Walt, 2010). Secondly and similar to poverty rates, African Americans are projected to be the most affected by growing unemployment rates in comparison to other American ethnic groups, and currently have an unemployment rate of 56%, doubling the national average (Dorsey, 2010; United States Department of Labor, 2010). Thirdly, the use of public assistance among African Americans is disproportionately high considering this is a minority racial group; however, the use is significantly lower than Caucasian, majority counterparts, according to the latest figures available (Bennett, 1995). Among these factors that promote underutilization of needed health care, African Americans are leading comprehensively.
According to a study assessing barriers preventing African Americans from seeking preventive screenings, particularly colorectal screenings, African Americans reported unawareness, fear, lack of transportation, lack of insurance, dislike for medical settings, inconvenience, apprehension of physicians, and indifference as barriers to screening (Good, Niziolek, Yoshida, & Rowlands, 2010). Lack of transportation and lack of insurance are indicators of low income and an unemployment status. In addition, the absence of economic stability fuels delays and hinders preventive measures like early detection and treatment of preventable diseases (Gelberg, 1996). Unemployment is a major contributor to economic instability and African Americans lead the nation in unemployment rates, as these rates among African Americans have increased from 12.1 to 16.2%, while the US unemployment rate is 10% (United Stated Department of Labor, 2010). The increased burden of unemployment among African Americans contributes to the lack of available resources need to maintain and manage one's health care needs. Lack of consistency in ascertaining care creates difficulty in chronic disease management (Runser, 2003); thus, contributing to poor health.
Conditions associated with unemployment, low income, and public assistance utilization, such as an uninsured/underinsured condition, disproportionately affect African Americans by approximately 20% more than Caucasians (Institute of Medicine Report, 2002; Schneider, Zaslavsky, & Epstein, 2002; National Center for Health Statistics, 2003; Weinick, Zuvekas, & Cohen, 2000). In a study reviewing medical records of 4,694 African Americans, health care settings for medically underserved had a higher prevalence and exhibited less control of diseases commonly plaguing communities; thereby, contributing to morbidities within this racial/ethnic group (Sheats et al., 2005). The restraint on health care access caused by low income reduces preventive service utilization and timely treatment.
Combined effects of low income and public assistance create a rich environment for delayed medical care. Low income and public assistance are likely a result of unemployment. The 'disconnect' between job existence and stability is often minimized or ignored (Runser, 2003). Understanding this disconnect brings clarity to the combined effect exacerbating poor health outcomes among African American populations vulnerable to an uninsured/ underinsured condition.
Prevalence of delayed medical care among homeless populations suggests homelessness as a contributing factor. During the 1980's, this population tripled in size due a decrease in the availability of affordable housing, a spike in crack/cocaine use, minimal mental health care coverage, and budget cuts at the federal level for public assistance (Kreider & Nicholson, 1997). The physical environment resulting from homelessness contributes to poor health. Homeless individuals heavily overuse the emergency departments at hospitals instead of clinics, and private medical services because they lack insurance and care needed is urgent (Schanzer, Dominguez, Shrout, & Caton, 2007). Health care among homeless populations is either emergent or nonexistent (Elliott, 2000). In a study of 974 homeless women residing in homeless shelters, 25% had a history of mental illness, ~50% were African American, ~33% had no usual source of care, ~ 50% were uninsured, and ~ 50% had a history of STIs (Gallagher, Andersen, Koegel, & Gelberg, 1997). These findings suggest the interaction of factors potentially related to health care utilization within homeless populations.
Unemployment, low income, and ethnicity has a causal association with health care usage (Runser, 2003). Given the correlation of these three variables to the health care access, these relationships were further explored among an at-risk African American population located in the Houston metropolitan are via the Cases and Places study sample. An assessment of a multivariate 'economic' effect on delayed medical care is applied to this population.
This surveillance study offers a secondary analysis to the cross-sectional data set collected through the "Cases and Places" study between 2004 and 2005 at the University of Texas, School of Public Health located in Houston, Texas. The questionnaire consisted of 119 questions and was administered to 215 research adults subjects, a majority African American population.
UT Health's Center for the Protection of Human Subjects approved the Cases and Places Study, HSC-SPH-03-085, which implemented a behavioral and environmental surveillance for gonorrhea (GC) transmission among high-risk populations (M. Ross, personal communication, December 1, 2006). Participants were either index patients with GC identified in two local clinics (n = 57 women) over a 12 month period, or men (n = 66) and women (n = 92) recruited at two local clubs/bars via a recruitment card. The two clinics and two bars were selected based on location in an effort to ascertain participant data representative of the Houston metropolitan area. Those recruited via the second method were later interviewed at a scheduled appointment with a researcher and offered the option to receive a GC test. Participants were consented and interviewed. Upon completion of the interview, incentives ($30-$40 for a 30-minute interview) were dispersed for time and designed to offset transportation or childcare costs. Standardized instructions were used in survey administration. The importance of the study, as well as the procedures in place to assure confidentiality, was explained to participants. There were potential subjects who refused to participate; however, data to calculate refusal rates were not recorded.
Measures: The categorical, outcome variable, 'delayed medical care', had a yes/no option inclusive of the following responses: don't know, refuse to answer, and not applicable. 'Delayed medical care' examined participant's tendency to delay medical care within the last 12 months.
Categorical, independent variables included gender, age, race, education, income, public assistance, and a history of homelessness.
Gender: Participants were identified as either women or men.
Age: Participants were included into one of four age categories: 18-19 years, 20-25 years, 26-30 years, or 31 years of age and older.
Race: Participants were categorized into one of four race options: African American, Hispanic or Latino, White, or other.
Education: Participants' education level completed was defined by one of four options: Elementary school, Junior high, High school or GED, or college.
Income: Income on a monthly basis, encompassing all sources of income, included two categories: none and $1-$3,001 and over.
Public assistance: Public assistance, assessing participant receipt of public assistance (inclusive of food stamps, TANF, SSI, etc.) had a yes/no response format.
A history of homelessness: A history of homelessness had a yes or no response format.
A frequency analysis performed on demographic variables included gender, age, race, education completed, income, and symptoms of depression (Table 1). Due to the 'Cases and Places study' methodology, there were more women (69.3%, n=149) than men (30.7%, n=66) (Table I). Mean (SD) age for the study sample was 34(11) and ranged from 18 to 62. African Americans represented 97.7% of the population. Completion of high school or obtaining a GED was the highest level of education for 72.6% of the study sample. In reference to income, most participants had a monthly household income of less than $1500 (73.1%) (Table I). Nearly 1/3 (28.8%) of the population reported public assistance use and 16/7% reported a history of homelessness.
A chi-square analysis of the relationship between economic determinants (income, public assistance, and a history of homelessness) and delayed medical care was assessed and stratified by 1) gender and 2) cases and places, cases only, and/or places only. Among cases and places, a significant correlation was noted between low income and delayed medical care among men (p < .01) (Table II). Similarly, a relationship between public assistance and delayed medical care was identified among men (p < .01). Lastly, a relationship was found among both genders of cases and places between a history of homelessness and delayed medical care (p < .05). Among both genders of places only, a significant relationship was noted between low income and delayed medical care (p < .05). Comparable to cases and places, the analysis of places, males specifically, revealed a significant correlation between public assistance and delayed medical care (p < .01). A significant relationship between a history of homelessness and delayed medical care was determined among both genders within the places group (p < .05). When stratified by gender, the correlation remained significant among women only (p < .05).
A binary logistic regression calculated an odds ratio (OR) and 95% CI for economic variables, collectively and individually, in relation to delayed medical care. Associations were stratified by 1) individual or collective analysis, 2) gender, and 3) cases and places, cases only, and/ or places only.
When assessed individually among cases and places, a protective effect to delayed medical care was identified among men who reported any source of income (OR = .09, .02-.53; Table III). Conversely, males in the places group, when analyzed as cases and places, were 16 times more likely to delay medical care (OR = 16.33, 1.67-159.75). Cases and Places with a history of homelessness were 2 times more likely to delay medical care (OR = 2.21, 1.04-4.71). There were no significant correlations among cases when assessed individually (Table III). Both genders in the places group reporting a source of income were less likely to delay medical care (OR = 0.23, .07-.81). Men and women designated as places with a history of homelessness were 3 times more likely to delay medical care (OR = 3.06, 1.33-7.08). The gender stratification revealed that women with a history of homelessness were also 3 times more likely to delay medical care (3.03, 1.03-8.89); thus, the correlation found among all places are likely due to the women in this group.
Collectively among cases and places, men in the place group who are on public assistance are 19 times more likely to delay medical care (OR = 19.25, 1.37-270.32). As in the individual assessment, there were no significant correlations among cases when economic determinants were assessed collectively. Like the individual assessment, the collective assessment of places revealed that both genders reporting an income source had a protective effect to delayed medical care (OR = .13, .03-.66), while the men in this group are likely the largely responsible for this significant association (OR = .03, 0.00-.48). Lastly, places with a history of homelessness were nearly 4 times more likely to delay medical care (OR = 3.55, 1.29-9.76). Significant gender differences were only revealed among men in the collective data analysis.
MAIN FINDING OF THE STUDY
The Cases and Places study was the first to utilize a behavioral surveillance method to assess this combination of economic determinants among a majority low income, minimally educated, African American population as it relates to delayed medical care. The study purpose is to determine whether low income, public assistance, and/or a history of homelessness are correlated to delayed medical care and to assess gender differences within the study population. Study findings suggest that among African American adults who delay medical care, the lack of financial means likely worsened efforts to access care.
The individual analysis revealing a significant correlation between low income and delayed medical care in the study population found in the chi-square analysis was clarified in the logistic regression analysis, where we identify that any form of income has a protective effect towards delayed medical care. Additionally, the collective analysis revealed a protective effect of income on delayed medical care among male places within the places only assessment.
WHAT IS ALREADY KNOWN ON THIS TOPIC
This correlation inversely reflects the disparity proposed by the US government, which acknowledges racial/ethnic minorities and low income populations in the US experience lower rates of insurance and access to health care as compared to counterparts(Agency for Healthcare Research and Quality, 2008; United States Department of Health & Human Services, 2010). African Americans with low income are > 25% more likely to be without a primary care physician compared to 20% of Whites (United States Department of Health & Human Services, 2010). Although the literature adequately addresses the racial disparity present regarding access to care(Brown, 2000; United States Department of Health & Human Services, 2010), gender stratification within racial/ethnic dynamics is not present.
WHAT THIS STUDY ADDS
Men are pillars in the African American communities; therefore, the need for financial assistance can be perceived as weak, resulting in a stigma detrimental to their reputation and status (Saluja et al., 2004). As patriarchs, African American men are responsible for the family; thus, delaying their own medical needs in order to secure the financial needs of the family is a cultural norm. The findings here suggest that maintaining some form of income, regardless of the amount, has a protective effect against the consequences of no income among African American men. Employment is a means of sustaining the integrity and esteem of the African American man while preventing delayed care for the patriarch of African American families. Further analysis of the correlation between financial means and willingness to preserve health among African American men is essential to future research.
Similar to the income assessment, a significant correlation between public assistance and delayed medical care was noted among men. In this case, African American men who use public assistance were 16 times more likely to delay medical care. When cases and places were analyzed collectively, the significant correlation between use of public assistance and delayed medical care persisted among male participants, as they were 19 times more likely to delay care. Use for public assistance reflects the need for bare essentials such as food and shelter; thus, timely access to care lacks the priority of a perceived emergency. Existing literature does not support this association, as the focus is on African American men leaving the family in order for their families to qualify for public assistance and have a chance at survival (Belgrave, 2009; Kaufman, 1997; Williams, 2007). Given the discontent between study findings and existing literature, this correlation among African American men should be explored further.
In regard to the significant correlation between a history of homelessness and delayed medical care among participants, African American men are more likely to be employed while homeless in comparison to their Caucasian counterparts (Levinson, 2004). It is suggested that African American men avoid homelessness for a longer period of time on little or no income in comparison to Caucasian counterparts (Levinson, 2004); thus, further supporting the protective effecting any form of income has on delayed medical care, as discussed earlier. Further exploration of the correlation of a history of homelessness and delayed medical care among places in the individual analysis reveal that women are largely responsible for this relationship. When assessed collectively, the significant correlation between a history of homelessness and delayed medical care among places persisted. Literature supports that African American women depend more on public assistance than their Caucasian counterparts while citing poverty or low income as their primary reason for their homeless situation (Levinson, 2004). Amidst poverty and dependence on public assistance, delayed medical care is a realistic outcome due to fixation on more immediate needs. A study of homeless African American women revealed that > 50% of the women were receiving public assistance and nearly 20% were employed (Levinson, 2004); thus, the public assistance is inefficient in providing the basic necessities of survival, such as shelter. Unmet health related needs among homeless adults (n = 966) include an inability to obtain needed 1) medical and surgical care, 2) prescription medications, 3) mental health care, 4) eye care, and 5) dental care (Baggett, O'Connell, Singer, & Rigotti, 2010). When basic essentials to survival are missing, the priority of timely access to care waivers; thus, delayed medical care becomes inevitable. It is important for researchers to discern what health related needs are unmet; however, it is more important, from a preventative standpoint, to understand the factors that give rise to this outcome. Predictors of unmet needs included inadequacy of food, out of home placement, and an uninsured status (Baggett, et al., 2010). The correlation between low income and a history of homelessness in relation to delayed medical care among African American women is apparent and should be researched further.
Future interventions would benefit from the Transtheoretical Model, purposed to evaluate one's current position on a behavior, while moving them towards the desired healthy behavior. In this case, the desired behavior would be seeking medical care at the onset of symptoms and moving towards the goal would be gradually minimizing delays to care (Banazak, 2000), in spite of financial restraints. Improving the perceived importance of timely care is needed as a focal point in future interventions, as health maintenance is vital to maintaining the ability to obtain the basic essentials of life.
LIMITATIONS OF THE STUDY
The Cases and Places study' size was small, limiting the power of the results and curtailed the likelihood of generalizability of findings to other African American urban populations. The study design was not randomized; thus, the external validity of the findings is limited. The existing literature assesses the economic determinants in light of race; however, gender stratification within ethnic/racial groups is not present in the literature and the comparison power of these findings is absent. The data was self reported, which has an inherent bias. The individual and collective designations for analysis was purposed to identify significant correlations potentially masked by misclassification or confounding bias.
Wide confidence intervals revealed in the logistic regression compromise the external validity of the findings related to those variables (ie. public assistance). Contributing factors to delayed medical care that could potentially confound the relationship between economic determinants and delayed medical care such as history of incarceration, secondary use of public assistance via someone who legally receives public assistance (ex: sharing food stamps), and employment information (ex: type of job/full time or part time status/number of jobs) were not observed in this study, but should be assessed in future studies.
FUTURE RESEARCH BASED ON THIS STUDY'S FINDINGS
According to the US government, 'access to quality care is vital to overall health and wellness' (United States Department of Health & Human Services, 2010). Given this fact, the data presented suggests economic factors such as low income, public assistance, and a history of homelessness places individuals, particularly African Americans who are demographically compatible to members of this study population, are at risk for compromised overall health and wellness. Processes, programs, and future research is needed to address the vitality of access to quality care in order to ensure this population has equitable access to overall health and wellness as compared to US citizens nationally.
Agency for Healthcare Research and Quality. (2008). National Healthcare Disparities Report.
American Cancer Society, Incorporated. (2009). Colorectal cancer rates higher in African Americans. Retrieved August 22, 2009, from http://www.cancer.org/doctor/NWS/content/ NWSIIxColorectalCancerRatesHigherinAfricansasp
Baggett, T. P., O'Connell, J. J., Singer, D. E., & Rigotti, N. A. (2010). The unmet health care needs of homeless adults: a national study. American Journal of Public Health, 100(7), 1326-1333.
Banazak, D. O. (2000). Minor depression in primary care. Journal of the American Osteopathic Association, 100, 783-787.
Belgrave, F. Z., Allison, K.W. (2009). African American Psychology: From Africa to America, Second Edition. Los Angeles: SAGE Publications.
Bennett, C. E., DeBarros, K. (1995). The Black Population in the United States: March 1994 and 1993. Washington, DC.
Brown, E. R., Ojeda, V.D., Wyn, R., and Levan, R. (2000). Racial and Ethnic Disparities in Access to Health Insurance and Health Care. Los Angeles, California: UCLA Center for Health Policy Research and Kaiser Family Foundation.
DeNavas-Walt, C., Proctor, B., Smith, J. (2010). Income, Poverty, and Health Insurance Coverage in the United States: 2009. Washington, DC: US Department of Commerce, Economics and Statistics Administration.
Dorsey, R. (2010). Unemployment highest among African Americans. Washington, DC: Howard University.
Elliott, M. (2000). The positive economic impact of health care for the homeless. Journal of the Medical Association of Georgia, 59(1), 15-19.
Gallagher, T. C., Andersen, R. M., Koegel, P., & Gelberg, L. (1997). Determinants of regular source of care among homeless adults in Los Angeles. Medical Care, 35(8), 814-830.
Gelberg, L. (1996). Homeless persons. In R. Anderson, Rice, T., & Kominiski, G. (Ed.), Changing the U.S. Health care System. San Francisco: Jossey-Bass.
Good, K., Niziolek, J., Yoshida, C., & Rowlands, A. (2010). Insights into barriers that prevent African Americans from seeking colorectal screenings: a qualitative study. Gastroenterology Nursing, 33(3), 204-208.
National Coalition for the Homeless. (2009). Minorities and Homelessness. Retrieved November 29, 2010, from http://www.nationalhomeless.org/factsheets/minorities.html
Kaufman, R. E. (1997). The cultural meaning of the "Welfare Queen": Using state constitutions to challenge child exclusion provisions. New York University Review of Law and Social Change, 23, 301-328.
Kellerman, A., Coleman, M. (2002). Care Without Coverage: Too Little, Too Late. Washington, DC: Institute of Medicine. Retrieved July 1, 2010, from http://www.iom.edu/~/media/Files/Report%20Files/2003/ Care-Without-Coverage-Too-Little-Too-Late/Uninsured2FINAL.pdf
Kreider, B., & Nicholson, S. (1997). Health insurance and the homeless. Journal of Health Economics, 6(1), 31-41.
United States Department of Labor. (2010). Employment Situation Summary. Washington, DC: Bureau of Labor Statistics. Retrieved November 24, 2010, from http://www.bls.gov/news.release/empsit.nr0.htm
Levinson, D. (2004). Encyclopedia of homelessness (Vol. 1): Sage Publications, Inc.
Institute of Medicine Report. (2002). Unequal treatment: confronting racial and ethnic disparities in health care Retrieved March 3, 2004, from http://www.iom.edu/reports.asp
Public Broadcasting Service. (2007). Homeless Facts and Figures. Retrieved November 29, 2010, from http://www.pbs.org/now/shows/526/homeless-facts.html
Runser, L., Eagen, S., and Olds, D. (2003). The Uninsured and Underinsured. Cleveland, OH: Case Western Reserve University.
Saluja, G., Iachan, R., Scheidt, P. C., Overpeck, M. D., Sun, W., & Giedd, J. N. (2004). Prevalence of and risk factors for depressive symptoms among young adolescents. Archives of Pediatrics & Adolescent Medicine, 155(8), 760-765.
Schanzer, B., Dominguez, B., Shrout, P. E., & Caton, C. L. (2007). Homelessness, health status, and health care use. American Journal of Public Health, 97(3), 464-469.
Schneider, E. C., Zaslavsky, A. M., & Epstein, A. M. (2002). Racial disparities in the quality of care for enrollees in medicare managed care. Journal of the American Medical Association, 257(10), 1288-1294.
Sheats, N., Lin, Y., Zhao, W., Cheek, D. E., Lackland, D. T., & Egan, B. M. (2005). Prevalence, treatment, and control of hypertension among African Americans and Caucasians at primary care sites for medically under-served patients. Ethnicity & Disease, 15(1), 25-32.
National Center for Health Statistics. (2003). Monitoring the nation's health. Early release of selected estimates from the National Health Interview Survey (NHIS). Atlanta, GA: Retrieved August 19, 2003, from http://www.cdc.gov/nchs/about/major/nhis/released200203.html
United States Department of Health & Human Services. (2010). Health Disparities: A case for closing the gap. Washington, DC: Retrieved December 5, 2010, from http://www.healthreform.gov/reports/ healthdisparities/
Weinick, R. M., Zuvekas, S. H., & Cohen, J. W. (2000). Racial and ethnic differences in access to and use of health care services, 1977 to 1996. Medical Care Research and Review, 57 Suppl 1, 36-54.
Williams, N. (2007). Where are the Men? The impact of incarceration and reentry on African American men and their children and families. Atlanta, GA: W.K. Kellogg Foundation.
Mandy Hill, DrPH, Assistant Professor, University of Texas Health, Medical School. Misha Granado, MPH, MS, University of Texas Health, Medical School. Jasmine Opusunju, MSEd, University of Texas Health, School of Public Health. Ronald Peters, DrPH, University of Texas Health, School of Public Health. Michael Ross, PhD, Dr.Med.Sc., University of Texas Health, School of Public Health. Corresponding Author: Mandy Hill, DrPH, 6431 Fannin JJL 420, Houston, TX 77030, email: email@example.com
Table I: Frequency Analysis of Demographic Variables Examined in the Cases and Places Study (n = 215) Variable Categories of Total Cases Places Variable N (%) N (%) N (%) Gender Women 149 (69.3%) 57 (100%) 92 (58.2%) Men 66 (30.7%) 0 (0%) 66 (41.8%) Age 18-19 12 (5.6%) 7 (12.3%) 5 (3.2%) 20-25 56 (26.0%) 30 (52.6%) 26 (16.5%) 26-30 25 (11.6%) 6 (10.5%) 19 (12.0%) 31 and older 122 (56.7%) 14 (24.6%) 108 (68.4%) Race African 210 (97.7%) 56 (98.2%) 154 (97.5%) American Hispanic or 1 (.5%) 1 (1.8%) 2 (1.3%) Latino White 3 (1.4%) 0 (0%) 1 (.6%) Other 1 (.5%) 0 (0%) 1 (.6%) Education Completed Elementary 1 (.5%) 0 (0%) 1 (.6%) school Junior high 31 (14.4%) 1 (1.8%) 30 (19.0%) High school 156 (72.6%) 46 (80.7%) 110 (69.6%) or GED College 27 (12.6%) 10 (17.5%) 17 (10.8%) Income (monthly) None 23 (10.7%) 12 (21.1%) 11 (7.0%) $1-$3,001 177 (82.3%) 43 (75.4%) 134 (84.8%) and over Public assistance Yes 62 (28.8%) 21 (36.8%) 41 (25.9%) No 153 (71.2%) 36 (63.2%) 117 (74.1%) History of homelessness Yes 36 (16.7%) 4 (7.0%) 32 (20.3%) No 179 (83.3%) 53 (93.0%) 126 (79.7%) Table II: Bivariate Analysis of economic-based independent variables as they relate to delayed medical care Independent Gender Pearson's p-value variable stratification [chi square] Income (1) 1.04 .31 Women .60 .44 Men 9.60 .00 * Public 1.42 .23 assistance (1) Women .03 .86 Men 9.16 .00 * History of 4.36 .04 * homelessness (1) Women 1.74 .19 Men 3.23 .07 Income (2) Women 1.46 .23 Public Women .05 .83 assistance (2) History of Women .09 .77 homelessness (2) Income (3) 6.01 .01 * Women .00 .97 Men 9.60 .00 Public 1.32 .25 assistance (3) Women .01 .93 Men 9.16 .00 * History of 7.26 .01 * homelessness (3) Women 4.28 .04 * Men 3.23 .07 - Places * -significant at the p < .05 level (1)--Cases and Places, (2)--Cases, (3)--places Table III: Logistics Regression on economic variables related to delayed medical care, stratified by case/ place status and gender INDEPENDENT VARIABLE TOTAL GENDER Men OR 95% CI OR INDIVIDUAL Income(1( .62 .25-1.56 .09 * Public assistance(1) 1.49 .77-2.88 16.33 * History of homelessness(1) 2.21 * 1.04-4.71 3.33 Income(2) -- -- -- Public assistance(2) -- -- -- History of homelessness(2) -- -- -- Income(3) .23 * .07-.81 -- Public assistance(3) 1.60 .71-3.60 -- History of homelessness(3) 3.06 * 1.33-7.08 -- COLLECTIVE Income(1) .52 .17-1.66 .03 * Public assistance(1) 1.41 .71-2.83 19.25 * History of homelessness(1) 2.09 .90-4.85 4.75 Income(2) -- -- -- Public assistance(2) -- -- -- History of homelessness(2) -- -- -- Income(3) .13 * .03-.66 .03 * Public assistance(3) 1.55 .64-3.75 19.25 * History of homelessness(3) 3.55 * 1.29-9.76 4.75 INDEPENDENT VARIABLE GENDER Women 95% CI OR 95% CI INDIVIDUAL Income1 .02-.53 1.67 .45-6.24 Public assistance(1) 1.67-159.75 1.07 .50-2.28 History of homelessness(1) .86-13.00 1.84 .74-4.61 Income(2) -- 2.68 .52-13.85 Public assistance(2) -- 1.14 .36-3.59 History of homelessness(2) -- .71 .07-7.30 Income(3) -- .95 .09-9.68 Public assistance(3) -- 1.05 .38-2.88 History of homelessness(3) -- 3.03 * 1.03-8.89 COLLECTIVE Income(1) .00-.48 1.47 .32-6.71 Public assistance(1) 1.37-270.32 1.06 .47-2.35 History of homelessness(1) .63-35.58 1.74 .65-4.63 Income(2) -- 3.41 .44-26.22 Public assistance(2) -- .99 .29-3.41 History of homelessness(2) -- .73 .07-8.07 Income(3) .00-.48 .68 .05-9.67 Public assistance(3) 1.37-270.32 1.14 .37-6.51 History of homelessness(3) .63-35.58 3.38 .99-11.56
|Gale Copyright:||Copyright 2011 Gale, Cengage Learning. All rights reserved.|